Tjitske van Engelen

93 Host response in patients with ICU-acquired pneumonia The majority of studies published on the association between immune changes and the development of secondary infections in the ICU focused on immune suppressive features of especially mononuclear cells, such as their reduced responsiveness to bacterial components, impaired antigen presentation capacity and signs of apoptosis [5–8]. Reduced HLA-DR expression on circulating monocytes and lymphocytopenia, both considered features of immune suppression, have been associated with an increased risk on nosocomial infections following sepsis or trauma [12–15]. Additionally, patients with ventilator-associated pneumonia showed lower CD4+ T cell counts and a reduced capacity of monocytes to release proinflammatory cytokines upon ex vivo stimulation, as compared with patients with non-respiratory nosocomial infections [16], and blood leukocyte gene expression profiles also suggested immune suppression in patients with ICU-acquired pneumonia [17]. Notably, while critical illness without doubt is associated with various immune suppressive reactions [5–8], patients admitted to the ICU concurrently show systemic hyperinflammatory responses which include cytokine release, and activation of the coagulation system and the vascular endothelium [6, 9, 18]. Our group recently reported that in ICU patients the degree of reduction in cytokine release by blood leukocytes, a common readout of immune suppression, was associated with simultaneously increasing systemic inflammation, stronger endothelial cell activation, loss of endothelial barrier integrity and enhanced procoagulant responses, suggesting that the strongest immune suppression occurs in those patients who concurrently display signs of stronger systemic inflammation [19]. In agreement, another study reported an inverse relationship between the plasma levels of IL-6, MMP-8 and CXCL9 (indicative of systemic inflammation) and the TNF production capacity of whole blood obtained from patients with sepsis [20]. Moreover, patients admitted for sepsis who later developed a secondary infection while in the ICU demonstrated greater aberrations in plasma biomarkers reflective of these hyperinflammatory pathophysiological domains relative to patients who did not develop an ICU-acquired infection [9]. The present investigation further supports the concept that secondary infections are associated with a widely disturbed immune response, encompassing distinct mediator systems, and characterized by not only immune suppression but also hyperinflammation. This notion is further reinforced by a study in trauma patients in whom multiple proinflammatory mediators were elevated within the first 24 hours after trauma in those who subsequently developed a nosocomial infection [21]. Likewise, in patients with sepsis elevated plasma midregional-proadrenomedulin levels were associated with an increased frequency of secondary infections [22]. Of note, differences in plasma biomarkers between cases and controls occurred despite similar disease severities at ICU admission. Changes in plasma host response proteins at study enrolment were most profound in patients in whom pneumonia was diagnosed relatively shortly (<5 days) or late (>10 days) after ICU admission. While the time windows chosen are arbitrarily, possibly these groups represent distinct pathobiological phenotypes, with in the first group changes that were partially already the consequence of an evolving infectious process in the lower airways, and in the latter group alterations with long-term impact on the susceptibility to pneumonia. To our knowledge such time-dependent analyses in the context of ICU-acquired infections have not been performed previously. Additional 5

RkJQdWJsaXNoZXIy MTk4NDMw